Table 3.

Backward regression analyses and multilevel analyses of the association between the outdoor childcare physical activity facilities and children's physical activity levels (number of observations = 689).

	Regression analysisa,b		Multilevel analysisa,c
	Regression coefficient (B)	P value	Regression coefficient (B)	P value
Age in years	.13	.019	—	—
Portable jumping equipment	.36	<.001	—	—
0 = not present, 1 = present
Portable slides	−.55	<.001	—	—
0 = not present, 1 = present
Fixed structured track	.53	<.001	.23	.003
0 = not present, 1 = present
Fixed sandbox	−.49	.002	—	—
0 = not present, 1 = present
Fixed swinging equipment	−.41	.001	—	—
0 = not present, 1 = present

^aVariables excluded from the analyses because they were present in either all or none of the childcare centers: portable equipment: ball equipment, push/pull toys and riding toys, and sand/water toys, twirling equipment; fixed equipment: merry-go-round, balancing surfaces.

^bVariables excluded from the final model because they were nonsignificant: portable equipment: climbing structures, floor play equipment; fixed equipment: climbing structures, see-saw, slides, and tunnels; background variables: sex.

^cFinal model comprises a first-order autoregressive (AR1) correlation for repeated measures within children. Other random effects were not included in the final model. Variables excluded from the final model because they were non-significant: portable equipment: climbing structures, floor play equipment, jumping equipment, and slides; fixed equipment: climbing structures, see-saw, slides, tunnels, sandbox, and swinging equipment; background variables: sex, age.